Split a /24 Into Four /26 Subnets

You have a /24 — say 10.0.10.0/24 — and you want four subnets out of it. Maybe one per environment (prod, stage, dev, mgmt), maybe one per floor in an office. This tutorial walks through the math once so you can do it in your head afterward.

---

What the slash means

A /24 is a network with 24 bits fixed for the network portion and 8 bits free for hosts. In 10.0.10.0/24:

• 10.0.10 is the network part (24 bits)
• the final octet is the host part (8 bits, or 256 possible values)

When you "split" a network, you are moving the boundary to the right. More bits for the network means more subnets, each with fewer hosts. Four subnets need 2 extra network bits (2^2 = 4), so the boundary moves from /24 to /26.

A /26 leaves 6 bits for hosts: 2^6 = 64 addresses per subnet. Minus the network address and the broadcast address, that is 62 usable hosts per subnet.

---

Step 1 — Calculate the four subnet boundaries

The step size is the subnet's block size: 256 / 4 = 64. So the subnets live at multiples of 64 in the final octet:

#	Network	Usable range	Broadcast
1	10.0.10.0/26	10.0.10.1 – 10.0.10.62	10.0.10.63
2	10.0.10.64/26	10.0.10.65 – 10.0.10.126	10.0.10.127
3	10.0.10.128/26	10.0.10.129 – 10.0.10.190	10.0.10.191
4	10.0.10.192/26	10.0.10.193 – 10.0.10.254	10.0.10.255

The quick way to verify: each subnet's broadcast address is one less than the next subnet's network address, and the last broadcast is .255 — the same broadcast as the parent /24. If those three checks line up, the split is correct.

---

Step 2 — The netmask in dotted-decimal form

If your equipment wants a mask instead of a prefix:

• /24 → 255.255.255.0
• /26 → 255.255.255.192

Where does 192 come from? A /26 sets the first 2 bits of the final octet to 1: 11000000 = 128 + 64 = 192. Same reasoning for other common prefixes:

• /25 → 10000000 → 128
• /26 → 11000000 → 192
• /27 → 11100000 → 224
• /28 → 11110000 → 240
• /29 → 11111000 → 248
• /30 → 11111100 → 252

You do not need to memorize the table. Count the ones, convert to decimal.

---

Step 3 — Assign them to purposes

A concrete split for a small office:

Subnet	Purpose	Gateway	Notes
10.0.10.0/26	Servers	.1	Static assignments only
10.0.10.64/26	Staff Wi-Fi	.65	DHCP pool .70–.126
10.0.10.128/26	Guest Wi-Fi	.129	Isolated, internet-only
10.0.10.192/26	Management	.193	Switches, APs, iDRAC/IPMI

A few practical notes:

• Pick .1 in each subnet as the default gateway so it is predictable. Some teams pick the highest usable (.62, .126, etc.) instead — either works, but pick one and stay consistent across subnets.
• Leave a small gap at the bottom of each subnet for statics before DHCP starts. If DHCP hands out .65 to a laptop, you cannot later put a printer on .65 without conflict.
• For guest Wi-Fi, make sure the firewall rules between subnets actually enforce the isolation. Putting guest traffic on a separate /26 does nothing on its own — it is a numbering convention until a rule backs it up.

---

Step 4 — Verify with a command

On Linux:

ipcalc 10.0.10.64/26

Output looks like:

Address:   10.0.10.64
Netmask:   255.255.255.192 = 26
Wildcard:  0.0.0.63
=>
Network:   10.0.10.64/26
HostMin:   10.0.10.65
HostMax:   10.0.10.126
Broadcast: 10.0.10.127
Hosts/Net: 62

Cross-check that against the table above. If the numbers disagree, you have the math wrong somewhere — trust the tool, redo the row.

If you prefer a web-based version, paste the CIDR into SysEmperor's CIDR Calculator — it shows the range, broadcast, and usable hosts without installing anything.

---

Going further: unequal subnets (VLSM)

The four-equal split is the simplest case. Real networks rarely want four equal pieces — usually one huge and a few small. You can split a /24 into one /25 and two /26s:

Subnet	Size	Range
10.0.10.0/25	126 usable	10.0.10.1 – 10.0.10.126
10.0.10.128/26	62 usable	10.0.10.129 – 10.0.10.190
10.0.10.192/26	62 usable	10.0.10.193 – 10.0.10.254

Rule of thumb for VLSM: always subnet from largest to smallest. Start with the biggest subnet at the beginning of the range, then bisect whatever is left. If you try to allocate small blocks first, you strand unusable gaps.

---

The only three facts worth memorizing

1. A new prefix bit halves the subnet.
2. Subnet block size = 2^(32 - prefix) addresses.
3. Usable hosts per subnet = block size − 2 (network + broadcast), except for /31 and /32, which are special cases used for point-to-point links and host routes respectively.

Everything else — masks, ranges, boundaries — falls out of those three.